plot_dimred: Plot a trajectory in a (given) dimensionality reduction
In dynplot: Visualising Single-Cell Trajectories

Description Usage Arguments Value Examples

Plot a trajectory in a (given) dimensionality reduction

plot_dimred(
  trajectory,
  color_cells = c("auto", "none", "grouping", "feature", "milestone", "pseudotime"),
  dimred = ifelse(dynwrap::is_wrapper_with_dimred(trajectory), NA,
    dyndimred::dimred_landmark_mds),
  plot_trajectory = dynwrap::is_wrapper_with_trajectory(trajectory) &&
    !plot_milestone_network,
  plot_milestone_network = FALSE,
  label_milestones = dynwrap::is_wrapper_with_milestone_labelling(trajectory),
  alpha_cells = 1,
  size_cells = 2.5,
  border_radius_percentage = 0.1,
  size_milestones = 6,
  size_transitions = 2,
  hex_cells = ifelse(length(trajectory$cell_ids) > 10000, 100, FALSE),
  grouping = NULL,
  groups = NULL,
  feature_oi = NULL,
  color_milestones = c("auto", "given", "cubeHelix", "Set3", "rainbow"),
  milestones = NULL,
  milestone_percentages = NULL,
  pseudotime = NULL,
  expression_source = "expression",
  arrow = grid::arrow(type = "closed", length = unit(0.1, "inches")),
  color_density = c("none", "grouping", "feature"),
  padding = 0.1,
  nbins = 1000,
  bw = 0.2,
  density_cutoff = 0.3,
  density_cutoff_label = density_cutoff/10,
  waypoints = dynwrap::select_waypoints(trajectory),
  trajectory_projection_sd = sum(trajectory$milestone_network$length) * 0.05,
  color_trajectory = "none"
)

`trajectory`	A dynwrap trajectory.
`color_cells`	How to color the cells. `"auto"`: Try to figure out how to color cells depending on whether one of the `grouping`, `feature_io`, `milestones` or `pseudotime` parameters are defined. `"none"`: Cells are not coloured. `"grouping"`: Cells are coloured according to a grouping (e.g. clustering). Either the `grouping` parameter or `trajectory$grouping` must be a named character vector. `"feature"`: Cells are coloured according to the values of a given feature (e.g. gene expression). Either the `expression_source` parameter or `get_expression(trajectory)` must be a matrix. Parameter `feature_oi` must also be defined. `"milestone"` (recommended): Cells are coloured according their position in the trajectory. The positioning of the cells are determined by parameter `milestone_percentages` or else by `trajectory$milestone_percentages`. The colours of the milestones can be determined automatically or can be specified by passing a tibble containing character columns `milestone_id` and `color` (See `add_milestone_coloring()` for help in constructing this object). `"pseudotime"`: Cells are coloured according to the pseudotime value from the root.
`dimred`	Can be A function which will perform the dimensionality reduction, see `dyndimred::list_dimred_methods()` A matrix with the dimensionality reduction, with cells in rows and dimensions (comp_1, comp_2, ...) in columns
`plot_trajectory`	Whether to plot the projected trajectory on the dimensionality reduction
`plot_milestone_network`	Whether to plot the projected milestone network on the dimensionality reduction
`label_milestones`	How to label the milestones. Can be TRUE (in which case the labels within the trajectory will be used), "all" (in which case both given labels and milestone_ids will be used), a named character vector, or FALSE
`alpha_cells`	The alpha of the cells
`size_cells`	The size of the cells
`border_radius_percentage`	The fraction of the radius that is used for the border
`size_milestones`	The size of the milestones
`size_transitions`	The size of the trajectory segments
`hex_cells`	The number of hexes to use, to avoid overplotting points. Default is FALSE if number of cells <= 10000.
`grouping`	A grouping of the cells (e.g. clustering) as a named character vector.
`groups`	A tibble containing character columns `group_id` and `color`. If `NULL`, this object is inferred from the `grouping` itself.
`feature_oi`	The name of a feature to use for colouring the cells.
`color_milestones`	Which palette to use for colouring the milestones `auto`: Determine colours automatically. If `color` is already specified in milestones tibble, this will be used. Otherwise, the colour scheme is determined by `milestone_palette_list$auto`. `given`: The `milestones` object already contains a column `color`. `cubeHelix`: Use the `rje::cubeHelix()` palette. `Set3`: Use the `RColorBrewer::brewer.pal(name = "Set3")` palette. `rainbow`: Use the `grDevices::rainbow()` palette.
`milestones`	Tibble containing the column `milestone_id` (character). If `color_milestones` is set to `"given"`, this tibble should also contain a column `color` (character), containing colour hex codes (e.g. `"#123456"`).
`milestone_percentages`	The milestone percentages.
`pseudotime`	The pseudotime from the root of the trajectory to the cells as a named numeric vector.
`expression_source`	Source of the expression
`arrow`	The type and size of arrow in case of directed trajectories. Set to NULL to remove arrow altogether.
`color_density`	How to color density, can be "none", "grouping", or "feature".
`padding`	The padding in the edges to the plot, relative to the size of the plot.
`nbins`	Number of bins for calculating the density.
`bw`	Bandwidth, relative to the size of the plot.
`density_cutoff`	Cutoff for density, the lower the larger the areas.
`density_cutoff_label`	Cutoff for density for labeling, the lower the further way from cells.
`waypoints`	The waypoints to use for projecting. Can by generated using `dynwrap::select_waypoints()`.
`trajectory_projection_sd`	The standard deviation of the Gaussian kernel to be used for projecting the trajectory. This is in the order of magnitude as the lengths of the milestone_network. The lower, the more closely the trajectory will follow the cells.
`color_trajectory`	How to color the trajectory, can be "nearest" for coloring to nearest cell, or "none".

A dimensionality reduction ggplot of the data.

data(example_bifurcating)
plot_dimred(example_bifurcating)


# plotting with umap
if (requireNamespace("uwot", quietly = TRUE)) {
  plot_dimred(example_bifurcating, dimred = dyndimred::dimred_umap)
}

# using a custom dimred
dimred <- dyndimred::dimred_mds(example_bifurcating$expression)
plot_dimred(example_bifurcating, dimred = dimred)

# coloring cells by pseudotime
plot_dimred(example_bifurcating, color_cells = "pseudotime")

# coloring cells by cluster
plot_dimred(
  example_bifurcating,
  color_density = "grouping",
  grouping = dynwrap::group_onto_nearest_milestones(example_bifurcating)
)